Errors due to data reduction in single-breath method for measurement of pulmonary blood flow.

The validity and precision of the single-breath method for determination of pulmonary blood flow suggested by Kim et al. (J. Appl. Physiol. 21: 1338-1344, 1966) have been estimated by a number of workers with very different results. The present work is an analysis of the errors that arise solely due to the data-reduction procedure used to calculate the pulmonary blood flow from the primary data. The analysis, which consists of a computer simulation of the system, shows that very considerable errors may arise from the data-reduction procedure itself and that these errors are very sensitive to small differences in the experimental design. The basic source of errors in the data-reduction procedure seems to be the use of a fitted polynomial to computer derivatives of a nonpolynomial experimental curve. Finally, the analysis suggests a design in which the errors would be quantitatively insignificant.